Moderated Mediation using PROCESS in SPSS, interpreting the output.

[Yuri]

Dear everybody,

Currently I'm working on some organisational research. Just to get a bit of a idea of what I'm research; I'm looking at impact of leadership on how enthusiastically work. I believe the relationship of the leader, and follower impact are important, as is the ambition of a worker.

So I'm looking at:
a. Leadership = TFL_MEAN = X
b. Enthusiasm = ENGAG_ME = Y
c. Relationship = LMX_MEAN = M
d. Ambition = Growth_N

Sorry for the jargon!

So I examine the impact of TFL (X) on ENGAG_ME (Y), where I take LMX_ME (M) as a mediator, and Growth Need  (W) as moderator of the relationship between X, and M. This gives the following model

So I use the PROCESS macro, using model 8, to instantly test the model (the output is below). So I'm wondering whether I'm interpreting the output right.

I Interpret it as that:
a. Growth_N is moderating the relationship between TFL_MEAN, and ENGAG_ME (due to the significant interaction, b =0.449, p=.028)
b. Significant direct effect of SL_MEAN on LMX_MEAN (b=.665, p <.001), but not on ENGAG_ME (b=.322, p = .109)
c. Growth_N is not moderating the relationship between SL_MEAN, and LMX_MEAN, due to the insignicant interaction (b=.098, p =.276)

So far, so good (I think). Next up are the "Conditional direct effect(s) of X on Y at values of the moderator(s)". I interpret this as (now I really don't know if this is right):
a. for low Growth_N (defined as minus 1 SD) there is an impact on ENGAG_ME, so when one has low levels the positive association between SL_MEAN, and ENGAG_ME strengthens.

And for the "Conditional indirect effect(s) of X on Y at values of the moderator(s):", I interpret this as (I don't really know whether this is right either):
a. The moderated mediation, and they are all significant, and their effect increases as their values increase. Thus the impact of SL_MEAN on ENGAG_ME, increases, as LMX_MEAN, and Growth_N increas.

After this the line "Indirect effect of highest order product:" pops up, showing LMX as insignificant. Does anyone know what this means?


Really sorry to bother you all with this load of SPSS output! I'm just really hoping there's a whizkid out here that can help me out!

Kindest Regards,

Yuri




**************************************************************************
Model = 8
    Y = ENGAG_ME
    X = SL_MEAN
    M = LMX_MEAN
    W = Growth_N
LMX MEAN
Sample size
        122

**************************************************************************
Outcome: LMX_MEAN

Model Summary
         R      R-sq       MSE         F       df1       df2         p
      ,693      ,481      ,133    36,442     3,000   118,000      ,000

Model
             coeff        se         t         p      LLCI      ULCI
constant     3,904      ,033   118,077      ,000     3,839     3,970
SL_MEAN       ,664      ,064    10,402      ,000      ,538      ,791
Growth_N      ,016      ,047      ,342      ,733     -,078      ,110
int_1         ,098      ,089     1,094      ,276     -,079      ,275

Product terms key:

 int_1    SL_MEAN     X     Growth_N

**************************************************************************
Outcome: ENGAG_ME

Model Summary
         R      R-sq       MSE         F       df1       df2         p
      ,558      ,311      ,675    13,209     4,000   117,000      ,000

Model
             coeff        se         t         p      LLCI      ULCI
constant     2,270      ,812     2,794      ,006      ,661     3,879
LMX_MEAN      ,629      ,207     3,034      ,003      ,218     1,039
SL_MEAN       ,322      ,199     1,617      ,109     -,072      ,716
Growth_N      ,348      ,107     3,258      ,001      ,137      ,560
int_2        -,449      ,202    -2,220      ,028     -,850     -,048

Product terms key:

 int_2    SL_MEAN     X     Growth_N

******************** DIRECT AND INDIRECT EFFECTS *************************

Conditional direct effect(s) of X on Y at values of the moderator(s):
  Growth_N    Effect        SE         t         p      LLCI      ULCI
     -,702      ,637      ,235     2,709      ,008      ,171     1,103
      ,000      ,322      ,199     1,617      ,109     -,072      ,716
      ,702      ,007      ,254      ,026      ,979     -,496      ,509

Conditional indirect effect(s) of X on Y at values of the moderator(s):

Mediator
          Growth_N    Effect   Boot SE  BootLLCI  BootULCI
LMX_MEAN     -,702      ,374      ,133      ,169      ,715
LMX_MEAN      ,000      ,418      ,132      ,208      ,733
LMX_MEAN      ,702      ,461      ,148      ,221      ,797

Values for quantitative moderators are the mean and plus/minus one SD from mean.
Values for dichotomous moderators are the two values of the moderator.

-----
Indirect effect of highest order product:

Mediator
            Effect  SE(Boot)  BootLLCI  BootULCI
LMX_MEAN      ,062      ,070     -,051      ,232

******************** INDEX OF MODERATED MEDIATION ************************

Mediator
             Index  SE(Boot)  BootLLCI  BootULCI
LMX_MEAN      ,062      ,070     -,051      ,232

******************** ANALYSIS NOTES AND WARNINGS *************************

Number of bootstrap samples for bias corrected bootstrap confidence intervals:
     1000

Level of confidence for all confidence intervals in output:
    95,00

NOTE: The following variables were mean centered prior to analysis:
 SL_MEAN  Growth_N

------ END MATRIX -----